The researchers stir-fried 50-gram batches of Sichuan pepper pericarp in a stainless-steel wok on a 2000-watt electric ceramic cooktop, preheated and held at 120 degrees Celsius as monitored by an infrared thermometer. This temperature was chosen as a controllable setting within the reported range for low-heat stir-frying. Samples were collected at one-minute intervals from zero to ten minutes, a sampling frequency selected to capture the critical changes occurring around the five-minute stage. For comparison, the team also processed pepper in a conventional iron wok over an open flame, following the traditional empirical endpoint of stir-frying until fragrant. The electric cooktop produced a more stable temperature profile with smaller fluctuations, indicating better process controllability than the open-flame method.

Scanning electron microscopy revealed the structural basis of the aroma release. The surface of unstir-fried pericarp was relatively smooth, covered by an intact waxy layer. After three minutes of heating, this waxy layer began to melt and rupture, micropores appeared on the surface, and oil was observed around the samples, corresponding to the sweating phenomenon familiar to processors. By five minutes, the surface structure had visibly collapsed and the number of pores had increased, which the authors suggest may have facilitated the release and diffusion of volatile compounds. Beyond eight minutes, charred fragments appeared on the surface, indicating severe tissue damage. These observations provide a structural explanation for the requirement that Sichuan pepper be stir-fried until fragrant, while also showing that excessive heating leads to pronounced tissue deterioration, aroma loss, and burnt off-odors.

Colorimetric measurements tracked the visible transformation. The lightness value, L*, declined continuously throughout stir-frying, while the red-green value, a*, increased initially, peaked at five minutes, and then decreased. The yellow-blue value, b*, fell steadily. The total color difference relative to the raw sample followed a corresponding trend. The turning point in a* after five minutes indicated that browning reactions, likely driven by the accumulation of Maillard reaction and caramelization products as well as the oxidative polymerization of phenolic compounds, intensified markedly beyond that stage. Cluster analysis and principal component analysis classified the samples into three distinct groups: zero to two minutes as under-stir-fried, three to six minutes as moderately stir-fried, and seven to ten minutes as over-stir-fried.

Headspace gas chromatography-mass spectrometry identified 26 volatile compounds, dominated by monoterpenes. In the raw material, sabinene, limonene, eucalyptol, myrcene, and 4-thujanol were the most abundant constituents. During stir-frying, most monoterpenes followed a pattern of initially increasing and subsequently decreasing: limonene, alpha-pinene, sabinene, and myrcene rose slightly within the first two minutes, dipped at three minutes, increased again to peak levels at four to five minutes, and then declined progressively. The researchers attribute the early increase to the melting and rupture of the waxy layer, which promoted desorption and release of terpenes from the matrix, while the later decline reflects evaporation and thermal degradation. Eucalyptol behaved differently, dropping sharply after three minutes and remaining low thereafter, despite having a boiling point similar to that of limonene. The authors suggest that its ether-oxide structure is more susceptible to thermal oxidation, and they propose eucalyptol as a potential chemical marker for monitoring the degree of stir-frying.

Relative odor activity value analysis revealed how the aroma-active profile was remodeled rather than simply reduced. In the raw pepper, 15 key aroma compounds were identified, with beta-terpineol showing the highest relative odor contribution, followed by limonene, linalool, and eucalyptol. From three minutes onward, limonene became the dominant contributor, maintaining the maximum relative odor activity value through the end of the process. Linalool declined from 61.69 in the raw sample to 35.13 at five minutes, while eucalyptol showed marked nonlinear changes, falling to 12.20 and 10.91 at four and five minutes before rebounding at six minutes. Myrcene remained consistently above the key aroma threshold throughout, ranging from 22.46 to 37.01. The number of key aroma components stayed within 15 to 16 across all time points, indicating that stir-frying altered the relative contribution of individual volatiles rather than uniformly diminishing overall aroma intensity.

To probe the molecular basis of aroma perception, the team performed molecular docking of seven key aroma compounds against seven olfactory receptor proteins retrieved from the AlphaFold3 database. Binding energies ranged from -4.646 to -7.317 kilocalories per mole, with the limonene-OR1A2 complex exhibiting the most favorable interaction. Limonene primarily formed hydrophobic contacts with residues Ala108, Tyr276, and Phe206, along with a pi-sigma interaction involving Tyr258. A 100-nanosecond molecular dynamics simulation using GROMACS confirmed the conformational stability of this complex: the system approached equilibrium after roughly 20 nanoseconds, the radius of gyration fluctuated narrowly between 2.08 and 2.20 nanometers, and the free energy landscape displayed multiple low-energy clusters. OR1A2 is one of the most broadly tuned members of the human olfactory receptor family and is known to be activated by various terpenoids, providing a potential molecular-level explanation for limonene’s prominent contribution to Sichuan pepper aroma.

Non-volatile constituents responded to heat in their own characteristic ways. Expressed on a dry-weight basis to account for progressive moisture loss, total flavonoids increased from 243.43 milligrams per gram in the raw sample to a maximum of 297.96 milligrams per gram at five minutes before declining, while total polyphenols fell continuously from 205.38 to 150.23 milligrams per gram at ten minutes. The two major numbing-related compounds, hydroxy-beta-sanshool and hydroxy-alpha-sanshool, both peaked at five minutes, at 7.47 and 7.58 milligrams per gram respectively, and subsequently decreased. Xanthoxylin and rutin also reached maxima during the early-to-middle stages before declining. Antioxidant activity followed a strikingly consistent trajectory: all four indices, including hydroxyl radical, DPPH, superoxide anion, and ABTS scavenging, increased and peaked at five minutes, with the hydroxyl radical scavenging rate nearly doubling from 33.80 to 65.73 percent, then falling significantly below raw-sample levels by ten minutes. The authors attribute the mid-stage enhancement to the accumulation of Maillard reaction products and the thermal release of phenolics from bound forms.

Correlation analysis tied these threads together. Partial least squares regression identified seven components, including total flavonoids, total polyphenols, eucalyptol, and gallic acid, as important contributors to the coordinated color-aroma changes, with variable importance in projection values exceeding 1.0. Spearman correlations showed that L* was positively associated with total polyphenols and gallic acid but strongly negatively correlated with 5-hydroxymethylfurfural, a Maillard reaction marker, consistent with progressive browning during prolonged heating. The a* value correlated positively with the numbing-related sanshools and xanthoxylin, as well as with multiple antioxidant indices. Sensory evaluation by eight trained panelists reinforced these links: sensory color correlated strongly with aroma and taste, sensory aroma correlated with xanthoxylin and ABTS capacity, and sensory taste correlated strongly with hydroxy-alpha-sanshool. Together, the integrated analyses demonstrated that chemical composition, aroma, color, and sensory perception transformed in a coordinated, nonlinear fashion during stir-frying.

The practical outcome is a reference processing window of four to five minutes at 120 degrees Celsius for standardized production, supported by the combined evidence from colorimetry, electronic-nose analysis, volatile profiling, HPLC determination of non-volatiles, antioxidant assays, sensory evaluation, and molecular simulation. The study also highlights the potential of the L* and a* color values as visual indicators of quality change, offering processors a practical tool for monitoring the transformation without relying on any single instrumental measurement. By showing that the spice’s flavor, color, and bioactivity rise and fall together within a matter of minutes, the research transforms an age-old culinary instinct, stir-fry until fragrant, into a quantitatively defined science of quality preservation.

Subject of Research: Chemical, aroma, and color changes in Sichuan pepper during stir-frying

Article Title: Synergistic changes and interrelationships among chemical composition, aroma, and color during the stir-frying of Sichuan pepper ( Zanthoxylum bungeanum maxim.)

Article References: Li, K., Wang, J., Zhang, X., Zhang, Q., Sun, J., & Zhao, C. (2026). Synergistic changes and interrelationships among chemical composition, aroma, and color during the stir-frying of Sichuan pepper (Zanthoxylum bungeanum maxim.). Food Chemistry: X, 40, Article 104644. https://doi.org/10.1016/j.fochx.2026.104644

Image Credits: AI Generated

DOI: 10.1016/j.fochx.2026.104644

Keywords: Sichuan pepper, Zanthoxylum bungeanum, stir-frying, volatile compounds, limonene, Maillard reaction, sanshools, flavonoids, antioxidant activity, colorimetry, molecular docking, food processing

The study, published in Food Chemistry: X, tackled a deceptively simple problem: what happens to porridge flavor when the millet it is made from has been sitting in storage? The answer required an unusually broad analytical arsenal. The researchers combined oxidation indicators, Maillard reaction markers, volatile profiling by gas chromatography–mass spectrometry, lipidomics, and metabolomics into a single integrated framework, allowing them to follow the molecular story from stored grain to finished porridge.

To compress years of storage into weeks, the team used accelerated oxidation. Fresh foxtail millet of the Jinmiao K1 variety, obtained from Xinzhou City in Shanxi Province, was sealed in polyethylene bags and stored in the dark at 50 degrees Celsius and 60 percent relative humidity. Using the Arrhenius relationship with a Q10 of 1.5, eight weeks under these conditions corresponds to roughly 22 weeks at room temperature. The researchers were careful to note that accelerated oxidation is not a fully equivalent substitute for normal storage, but it provides a mechanistic window into the reactions that accumulate over time.

At two-week intervals, samples were split in two: one portion ground raw, the other cooked into porridge with water at a 1:30 ratio for 30 minutes, then freeze-dried for analysis. The chemical readouts were striking. Malondialdehyde, a hallmark of lipid oxidation, rose in raw millet from 0.5669 to 0.6288 milligrams per kilogram over eight weeks, while cooked porridge samples carried far higher levels, climbing from 1.1616 to 1.3494 milligrams per kilogram. Peroxide values, which track early-stage oxidation, also increased during storage, from 0.6675 to 1.4155 milliequivalents per kilogram in raw grain. Cooking told a more complicated story: peroxides formed during heating but simultaneously decomposed, so the net effect depended on how long the grain had been stored beforehand.

Proteins were not spared. Carbonyl content, a classic marker of protein oxidation, increased in raw millet from 22.79 to 24.01 nanomoles per milligram of protein, and jumped dramatically after cooking, reaching 29.11 nanomoles per milligram by week eight. Free amino groups fell in raw millet as oxidation and Maillard-type reactions consumed them, but rose after cooking, suggesting heat-induced protein hydrolysis partly offset the losses. Total sulfhydryl groups dropped during storage as exposed thiols oxidized into disulfide bonds, yet increased after cooking, likely reflecting protein unfolding and disruption of disulfide structures. Surface hydrophobicity declined in both raw and cooked samples, indicating that hydrophobic regions became progressively buried through oxidation, Maillard chemistry, and heat aggregation.

Amino acid oxidation products told a parallel tale. Dityrosine, formed by oxidative crosslinking of tyrosine residues, and kynurenine and N′-formylkynurenine, products of tryptophan degradation, all increased during storage and rose further after cooking. The team interpreted this as evidence that heating expands the protein structure, exposing aromatic amino acid residues to oxidation. Meanwhile, soluble fluorescent advanced glycation end products—hallmarks of protein-sugar crosslinking—climbed steadily in both raw and cooked samples, from roughly 302 to 359 units per microgram of protein in raw millet, confirming continuous Maillard reaction activity throughout storage and cooking.

Reducing sugars, the other half of the Maillard equation, declined during storage in both sample types, from 4.69 to 4.47 milligrams per gram in raw millet, consistent with their consumption in AGE formation. 5-Hydroxymethylfurfural, an intermediate Maillard marker, actually fell in raw millet from 7.05 to 0.32 micrograms per gram, a pattern the researchers attributed to protein competition: free amino groups capture Maillard intermediates like 3-deoxyglucosone, blocking 5-HMF formation, while preexisting 5-HMF is consumed in late-stage reactions with proteins. In fresh millet, cooking decreased 5-HMF further, likely because heating exposes amino residues that trap it into undetectable adducts.

The volatile compounds—the molecules we actually smell—changed markedly. After cooking, acids and aldehydes dominated the profile, indicating that heating intensified both lipid oxidation and Maillard reactions. Screening by odor activity values and multivariate statistics identified 20 key volatiles during storage. Nonanal, 2-pentylfuran, 1-octen-3-ol, 2-heptanone, and pentanal traced back to lipid oxidation, while benzaldehyde, maltol, acetophenone, and 3-methylbutanal arose from Maillard and Strecker chemistry. After cooking, compounds like (E,E)-2,4-decadienal and (E)-2-octenal emerged as lipid thermal degradation products. Venn analysis revealed 10 storage-specific, 9 cooking-specific, and 10 shared volatiles, with shared compounds like hexanoic acid, decanal, and nonanal generally increasing after cooking—evidence that heating triggered volatile formation from precursors already altered by storage.

Lipidomics identified 181 lipids across seven major classes, with glycerolipids making up over half. Volcano plots showed progressively more differential lipids with storage time: 4, 17, 25, and 32 lipids changed significantly at 2, 4, 6, and 8 weeks in raw millet, predominantly down-regulated. Oxidation derivatives like FA 18:1;O3 and FA 18:2;O3 accumulated while membrane phospholipids like PA 36:4 declined, indicating ongoing hydrolysis and oxidation driven by lipase and reactive oxygen species. After cooking, the pattern shifted: free fatty acids dropped while oxygenated derivatives rose, reflecting accelerated thermal oxidation. Metabolomics identified 108 metabolites, dominated by fatty acids, carbohydrates, and amino acids. Sixteen metabolites changed in both raw and cooked samples, including trehalose, galactose 1-phosphate, gluconic acid, and oxidized fatty acids like 9-HODE and 13-HOTrE—suggesting that the same precursor-conversion pathways operate in stored grain and porridge alike.

Correlation networks tied the molecular threads together. Oxidized fatty acids like FA 18:2;O2 were positively associated with 3-methylbutanal, acetophenone, hexanoic acid, and benzaldehyde, while gluconic acid showed the highest connectivity in the metabolite-volatile network, linked to acetophenone, 3-methylbutanal, hexanoic acid, and 2-pentylfuran. The team built a conceptual diagram summarizing how lipid oxidation, protein oxidation, and Maillard reactions may jointly shape the volatile profile of porridge. The findings provide a chemical basis for understanding storage-related quality changes in foxtail millet porridge and may guide future studies under realistic storage conditions. For a grain that has sustained civilizations, the science of its flavor is finally catching up with its history.

Subject of Research: Storage-driven lipid oxidation, protein oxidation, and Maillard chemistry reshaping foxtail millet porridge volatiles

Article Title: Effects of accelerated foxtail millet storage on porridge volatiles: An integrated multi-omics study

Article References: Wen, R., Liang, Y., Qiao, J., Bai, Y., Yu, L., Niu, W., Zhao, S., & Zhang, L. (2026). Effects of accelerated foxtail millet storage on porridge volatiles: An integrated multi-omics study. Food Chemistry: X, 40, Article 104606. https://doi.org/10.1016/j.fochx.2026.104606

Image Credits: AI Generated

DOI: 10.1016/j.fochx.2026.104606

Keywords: foxtail millet, porridge, food chemistry, lipid oxidation, Maillard reaction, volatiles, lipidomics, metabolomics, storage quality, flavor chemistry, Effects, accelerated

The study, published in Food Chemistry: X, evaluated both hydrolyzed and unhydrolyzed tara pod phenolic extracts in a bovine serum albumin–methylglyoxal model system and in real quinoa and cañihua cookies baked at 180 °C for 16 minutes. The researchers chemically characterized the extracts using UPLC–PDA–QTOF–MS and found that acid hydrolysis progressively converted gallotannins, specifically polygalloylquinic acids, into free gallic acid. After 20 hours of hydrolysis, the gallic acid content reached 833 mg per gram of dry weight, compared with only 6 mg in the unhydrolyzed extract. Increasing the degree of hydrolysis was associated with higher total phenolic content and greater antioxidant capacity, attributed to structural modifications that increase the availability of phenolic hydroxyl groups with radical-scavenging ability.

In the BSA–MGO model system, which bypasses the initial conversion of reducing sugars into reactive carbonyl compounds, tara extracts exhibited marked concentration-dependent antiglycation activity. At the highest concentration tested, 10 mg gallic acid equivalents per milliliter, the unhydrolyzed tara extract showed the numerically highest inhibition of fluorescent AGEs formation at 90.3 percent, comparable to the prototype inhibitor aminoguanidine, which reduced fluorescence by approximately 82 percent. Interestingly, at lower concentrations, inhibition tended to increase with the degree of hydrolysis, suggesting that the antiglycation activity depends on both concentration and structural characteristics of the phenolic compounds. The researchers also noted that polyphenols can interact with proteins through covalent and non-covalent bonds, forming complexes that may decrease the accessibility of amino groups and reduce their reactivity in the OPA assay, complicating interpretation of the results.

When the extracts were incorporated into real cookies at a fixed concentration of 10 mg gallic acid equivalents per 100 g flour, the effects on acrylamide and AGEs formation were strongly dependent on both the food matrix and the degree of hydrolysis. Supplementation with the unhydrolyzed tara extract resulted in the lowest acrylamide levels in both quinoa and cañihua cookies, followed by the extensively hydrolyzed extract. In contrast, extracts with intermediate degrees of hydrolysis were less effective and, in some cases, produced acrylamide concentrations similar to or higher than those of control cookies. The researchers attributed this non-linear response to hydrolysis-induced changes in the phenolic profile, noting that gallotannins interact with proteins and amino acids through hydrogen bonds and hydrophobic association, while interactions with sugars are restricted to hydrogen bonding, and that compounds with higher numbers of galloyl residues provide higher binding rates to proteins and sugars.

Notably, acrylamide levels in quinoa and cañihua cookies were lower than those detected in wheat cookies prepared with the same formulation, which showed 64.1 μg per kg dry weight. All treatments remained below the EFSA reference level of 265 μg per kg for cookies and related products, and far below commercial cookies, which can contain approximately 1340 μg per kg. The comparatively low acrylamide formation in pseudocereal cookies may relate to their lower asparagine content, given the key role of this amino acid as an acrylamide precursor during thermal processing in cereal-based bakery products.

Regarding fluorescent AGEs, supplementation with tara extracts did not significantly reduce their formation in either quinoa or cañihua cookies compared with controls. The researchers suggested that endogenous phenolic compounds, including flavonoids such as rutin, in both pseudocereal matrices may already exert protective effects through sequestration of reactive dicarbonyl compounds, reducing the additional benefit of tara-derived phenolics. Furthermore, fluorescence-based determination of AGEs may be affected by interference from non-AGEs fluorophores present in complex food matrices, limiting the specificity of this marker. In contrast, protein-bound AGEs, namely Nε-carboxymethyl-l-lysine and Nε-carboxyethyl-l-lysine, showed a response pattern more similar to acrylamide and was strongly dependent on both the food matrix and the degree of extract hydrolysis. In quinoa cookies, the unhydrolyzed extract significantly reduced both CML and CEL concentrations, whereas extracts with higher degrees of hydrolysis were less effective and, in some cases, even increased protein-bound AGEs levels. In cañihua cookies, both unhydrolyzed and partially hydrolyzed extracts reduced CML, while 24-THE and 57-THE extracts significantly lowered CEL formation.

Given its greater mitigating effect on both acrylamide and protein-bound AGEs, the unhydrolyzed tara extract was selected for further evaluation at different concentrations, 2.5, 10, and 15 mg gallic acid equivalents per 100 g flour. In cañihua cookies, significant reductions in fluorescent AGEs and CEL were observed at intermediate supplementation levels, while a significant reduction in CML was observed at the lowest concentration. Acrylamide formation in cañihua cookies showed a non-linear concentration-dependent response, with the lowest levels at intermediate supplementation levels, 42.0 and 36.1 μg per kg dry weight at 2.5 and 10 mg, compared with 67.0 μg per kg in the control. In quinoa cookies, however, UTE supplementation did not significantly mitigate acrylamide formation, and increasing the supplementation level resulted in progressively higher acrylamide concentrations, reaching 60.5 μg per kg at the highest dose. These contrasting responses indicate that the antiglycation effect of the unhydrolyzed extract in complex food matrices is both matrix- and concentration-dependent and does not follow a simple linear dose–response relationship.

The physicochemical changes induced by supplementation further highlighted the complexity of the system. Progressive reduction in cookie moisture content with increasing extract concentrations was attributed to a modifying effect of gallotannins on water molecular mobility, which influences the availability of Maillard precursors during baking. In quinoa cookies, the moisture reduction coincided with a gradual increase in acrylamide levels, but this pattern was not observed in cañihua cookies, suggesting that moisture variations, while influential, were not the only determinant factor. Differences in total and soluble protein contents, reducing sugar concentrations, and endogenous phenolic profiles between the two Chenopodium species likely generate different chemical environments during baking, modulating both AGEs formation and the effect of supplementation. Cañihua cookies exhibited higher total protein content, 9.44 g per 100 g dry weight compared with 7.41 g in quinoa cookies, while quinoa cookies showed slightly higher reducing sugar concentrations and higher levels of both early and advanced Maillard reaction products.

The study concludes that tara pod phenolic extracts show the capacity to mitigate acrylamide and AGEs formation in quinoa and cañihua cookies, although the magnitude and direction of the effect is highly dependent on the food matrix, extract concentration, and the specific Maillard reaction marker evaluated. The different responses observed for acrylamide, fluorescent AGEs, CML, and CEL indicate that their formation is influenced by complex interactions between tara-derived polyphenols and endogenous matrix components during baking, as well as by physicochemical changes induced by supplementation. These findings highlight tara pods as a promising source of phenolic extracts capable of modulating thermally induced reactions in bakery products based on Andean pseudocereals, although their effects cannot be generalized as uniform mitigators of acrylamide and AGEs formation. The researchers emphasize that incorporating tara phenolic extracts modifies the bioactive and physicochemical characteristics of the final product, and that these modifications should be taken into consideration during the formulation of bakery products enriched with natural phenolic extracts.

Subject of Research: Food Chemistry

Article Title: Modulation of acrylamide and advanced glycation end products in quinoa and cañihua cookies: Matrix and marker dependent effects of hydrolyzed and unhydrolyzed tara pod phenolic extracts

Article References: Aguilar-Galvez, A., García-Ríos, D., Meléndez-Gamarra, A., Quispe, T., Calsin-Cutimbo, M., Chirinos, R., Pedreschi, F., Pedreschi, R., & Campos, D. (2026). Modulation of acrylamide and advanced glycation end products in quinoa and cañihua cookies: Matrix and marker dependent effects of hydrolyzed and unhydrolyzed tara pod phenolic extracts. Food Chemistry: X, Article 104614. https://doi.org/10.1016/j.fochx.2026.104614

Image Credits: AI Generated

DOI: 10.1016/j.fochx.2026.104614

Keywords: acrylamide, advanced glycation end products, Maillard reaction, quinoa, cañihua, tara pod, phenolic compounds, food safety, bakery products, Andean pseudocereals, Modulation, advanced

The study, published in Food Chemistry: X, set out to resolve a stubborn trade-off that has plagued previous attempts to make fried meat safer. Researchers led by Wei Quan of Changsha had earlier identified chitosan, a linear polysaccharide obtained by deacetylating chitin from crustacean shells, as a remarkably effective inhibitor of heat-induced hazards in meat. Chitosan works because its abundant free amino groups form Schiff bases with reactive carbonyl intermediates, intercepting the molecular building blocks of both HAs and AGEs before they can assemble into their final, dangerous forms. But there was a catch: at the concentrations needed for meaningful inhibition, plain chitosan imparted a pronounced off-flavor to the finished product, a sensory penalty that no amount of safety improvement could justify commercially. The new work asked a deceptively simple question: if you change the chemistry of chitosan, can you change what it attacks, and what it leaves alone?

To answer it, the team compared five structurally distinct chitosans in a real food matrix rather than a simplified model system. Alongside unmodified chitosan, they tested carboxyl-modified chitosan, sulfhydryl-modified chitosan, quaternary ammonium-modified chitosan, and carboxymethyl-modified chitosan. Each modification was chosen for a specific mechanistic hypothesis. Carboxyl groups were expected to chelate pro-oxidant metal ions that drive oxidative steps in the Maillard reaction. Thiol groups, being strongly nucleophilic, were predicted to trap reactive dicarbonyl intermediates such as glyoxal and methylglyoxal through direct nucleophilic addition. Permanent positive charges on the quaternary ammonium derivative were expected to electrostatically disrupt negatively charged Maillard intermediates. Carboxymethylation, prized for water solubility, was suspected of a hidden cost: it masks the very amino groups that make chitosan work.

The experimental design was refreshingly practical. Fresh beef longissimus lumborum was ground, mixed with 0.1 percent by weight of each chitosan derivative as a dry powder, marinated for thirty minutes, and shaped into patties. The patties were then air-fried at 200 degrees Celsius for ten minutes per side, conditions typical of both home cooking and laboratory studies of contaminant formation. After cooking, the researchers extracted and purified the HAs and AGEs from the meat and quantified them using ultra-high-performance liquid chromatography coupled to triple quadrupole mass spectrometry, targeting four representative HAs, namely Harman, Norharman, PhIP, and MeIQx, and the two major AGEs, carboxymethyllysine and carboxyethyllysine.

The results revealed a striking structure-activity relationship. All five treatments significantly reduced both HAs and AGEs compared with untreated controls, but the magnitude and selectivity of inhibition varied dramatically. Unmodified chitosan delivered broad-spectrum suppression, cutting Norharman by 74.8 percent, PhIP by 68.6 percent, and MeIQx by 66.1 percent, with more modest reductions of roughly 22 to 24 percent in the two AGEs. The thiolated derivative, however, emerged as the star performer, achieving the strongest and broadest inhibition: MeIQx fell by 75.0 percent, Harman by 65.1 percent, carboxymethyllysine by 26.9 percent, and carboxyethyllysine by 32.0 percent. Critically, it was the only derivative that simultaneously topped the rankings for both IQ-type HAs and AGEs, suggesting that thiol groups excel at intercepting the reactive intermediates shared by both hazard pathways.

The other derivatives told more selective stories. Carboxyl-modified chitosan was the champion against PhIP, reducing it by 78.4 percent, consistent with the idea that metal chelation preferentially disrupts the oxidative steps crucial to that particular HA’s formation, yet it showed only moderate effects on the beta-carbolines. The quaternary ammonium derivative displayed the opposite pattern, achieving the strongest inhibition of beta-carbolines while performing worst of all against carboxyethyllysine, evidence that electrostatic disruption is a pathway-selective tool rather than a universal weapon. Carboxymethylated chitosan fared worst across the board, and the reason is chemically elegant: substituting the free amino groups with bulky carboxymethyl moieties destroys the Schiff base chemistry that underpins carbonyl scavenging, and no amount of incidental metal-chelating capacity could compensate. The lesson is that direct trapping of reactive carbonyls, not indirect antioxidant activity, is the dominant mechanism controlling these contaminants.

But hazard inhibition is only half the story, and this is where the study breaks new ground. The researchers went on to measure cooking loss, water distribution, color, texture, lipid oxidation, and protein oxidation, and then profiled the volatile aroma compounds using gas chromatography-ion mobility spectrometry, a technique sensitive enough to fingerprint the subtle differences between treatments. The water data showed that chitosan and its thiolated and carboxymethylated forms significantly reduced cooking loss and shifted water from the free to the immobile fraction, a change associated with juicier, more stable meat. The thiolated derivative also proved the most effective protector of proteins, cutting total carbonyl content from 2.25 to 0.93 nanomoles per milligram, likely by scavenging reactive oxygen species and forming protective mixed disulfides with protein thiols.

The flavor analysis delivered the study’s most commercially significant finding. The thiolated chitosan altered the volatile profile least of all treatments: the levels of hexanal, heptanal, octanal, and nonanal, the straight-chain aldehydes responsible for the fatty, grassy, and roasted notes of cooked meat, remained closest to those of the untreated control. The explanation lies in chemical selectivity. Thiol groups react preferentially with the highly reactive dicarbonyl intermediates that serve as HA and AGE precursors, while sparing the less reactive aldehydes that carry the aroma. By contrast, unmodified chitosan and the quaternary ammonium derivative depleted the desirable aldehydes indiscriminately through non-specific Schiff base formation and electrostatic interactions respectively, potentially stripping the patties of their characteristic meaty aroma even as they made them safer.

Multivariate statistics tied the whole picture together. Principal component analysis of the combined hazard and flavor dataset showed that the first two components captured 74.8 percent of the total variance, and the spatial map placed the thiolated derivative closest to the control along flavor dimensions while farthest from the hazard loadings, a quantitative confirmation of its dual advantage. Partial least squares discriminant analysis identified eleven key volatile compounds, predominantly aldehydes, as the primary drivers of differentiation between treatments. Taken together, the data establish a coherent framework: the accessibility and reactivity of nucleophilic groups determine both how well a chitosan derivative blocks heat-induced contaminants and how gently it treats the flavor chemistry of cooking meat.

The authors are careful to note the limits of their work. The proposed mechanisms involving glyoxal and methylglyoxal are inferred from inhibition patterns and supported by separate model-system studies rather than by direct measurement of intermediates in the meat itself, and no human sensory panel has yet confirmed that the flavor differences detected instrumentally translate into perceptible changes for consumers. Future work, they suggest, should include trapping assays, spectroscopic validation, consumer taste tests, and extension to other foods and cooking conditions. Even so, the implications are considerable. A food-grade, biocompatible additive derived from shellfish waste, added at just 0.1 percent, could meaningfully reduce two classes of potentially harmful compounds in one of the world’s most popular cooked foods while leaving its aroma essentially intact. As demand grows for healthier processed meats that do not taste like a compromise, the humble thiol may prove to be the food industry’s most useful new chemical handle.

Subject of Research: Structure-dependent inhibition of heterocyclic amines and advanced glycation end products by chemically modified chitosans in air-fried beef patties

Article Title: Structure-dependent inhibition of heterocyclic amines and advanced glycation end products by chemical modified chitosans in air-fried beef patties: Balancing safety and flavor quality

Article References: Quan, W., Liu, J., Liu, S., Zhou, L., Wu, Z., Lou, A., Liu, Y., Shen, Q., Zeng, M., Chen, J., Xue, C., Yang, D., & Mo, L. (2026). Structure-dependent inhibition of heterocyclic amines and advanced glycation end products by chemical modified chitosans in air-fried beef patties: Balancing safety and flavor quality. Food Chemistry: X, Article 104620. https://doi.org/10.1016/j.fochx.2026.104620

Image Credits: AI Generated

DOI: 10.1016/j.fochx.2026.104620

Keywords: chitosan, heterocyclic amines, advanced glycation end products, air frying, Maillard reaction, food safety, beef patties, thiol modification, flavor chemistry, GC-IMS, food chemistry, meat processing

The research team, led by Shaohong Xu and colleagues, built a simplified but rigorous experimental system: casein, the predominant protein in milk, was mixed with glucose at a fixed concentration and incubated at 90 degrees Celsius for up to 24 hours. Into this mixture they added either sodium dihydrogen phosphate, abbreviated MSP, or disodium hydrogen phosphate, DSP, at low and high doses of 10 and 100 millimolar. Crucially, the pH of every sample was adjusted to 7.0 before the reaction began, a design choice that allowed the researchers to isolate the intrinsic chemical effects of the phosphate species and ionic strength from the well-known influence of acidity itself. Samples were then pulled at eight time points, from half an hour to a full day, and interrogated with an arsenal of analytical techniques.

The most visually dramatic result came from simple colorimetry. After 24 hours of heating, the browning index of the untreated control casein reached about 28 units. Adding low-dose MSP slashed that figure to roughly 4.5, producing a nearly white sample, while high-dose MSP still kept browning well below the control at about 17 units. DSP did the opposite: low-dose DSP pushed the index to about 41, and high-dose DSP sent it soaring to nearly 222, an eightfold increase over the control, accompanied by a deep reddish-brown hue. The explanation lies in acid-base chemistry. MSP is a weakly acidic phosphate that pulls the local reaction environment toward lower pH values unfavorable to the Maillard cascade, whereas DSP, a weakly alkaline salt, nudges the local pH upward and thereby facilitates the initial attack of glucose on the protein’s amino groups.

Spectroscopic measurements traced the reaction’s progress through its intermediate and advanced stages. Ultraviolet absorbance at 294 nanometers, a proxy for early Maillard intermediates such as furfurals and reductones, rose steadily in all samples over 24 hours, but the 100 millimolar phosphate groups, both MSP and DSP, accumulated significantly more intermediates than the control throughout the run. Absorbance at 420 nanometers, which tracks the colored melanoidin pigments of the reaction’s late stage, told a complementary story: every phosphate-supplemented sample browned faster than the control, and the effect was clearly concentration-dependent, with the high-dose groups reaching the highest pigment levels. Even the low-dose groups, though weaker, remained measurably above the control, hinting that ionic strength itself can redirect reaction pathways and blunt pigment formation efficiency.

Beneath the color changes, the phosphates were quietly remodeling the protein itself. Dynamic light scattering revealed that the untreated control casein underwent dramatic particle growth during glycation, a signature of aggregation and conformational unfolding. Phosphate-treated samples, by contrast, showed far gentler increases in particle size, and the effect scaled with dose: 100 millimolar phosphate consistently yielded smaller particles than 10 millimolar. This mirrors earlier observations in meat and dairy systems, where phosphates reduce protein diameter and boost solubility through charge shielding and dispersion mechanisms. In practical terms, phosphate addition kept the glycated protein dispersed rather than clumped, a property that also improved water solubility and emulsifying capacity, traits highly prized in beverage and encapsulation applications.

The salts also unmasked the protein’s reactive chemistry. Using the OPA assay to quantify free amino groups, the researchers found that at the very start of the reaction, before any glycation had proceeded, all phosphate-treated samples displayed significantly higher free amino content than the control, with the high-dose DSP group recording the highest value. This indicates that phosphate pretreatment exposes lysine and arginine residues that are normally buried inside the folded protein, handing the glucose molecule more accessible docking sites. Over the first two hours, the phosphate groups showed dynamic shifts, with some samples consuming free amino groups early while the high-dose MSP group sustained elevated levels, demonstrating persistent reactive capacity. Phosphate, in other words, acts not merely as a dispersant but as a molecular chaperone of sorts, rearranging the protein to make it more reactive.

Circular dichroism spectroscopy confirmed that this rearrangement was structural and remarkably fast. Far-UV spectra collected at one, two and 16 hours showed that glycated samples shifted from a highly ordered, beta-sheet-rich conformation toward a more flexible, disordered state. Alpha-helix content rose modestly from about 6 percent in the control to 8 or 9 percent in treated samples, while parallel beta-sheet structure plummeted by 10 to 15 percentage points, and beta-turn and random coil fractions climbed correspondingly. Strikingly, the secondary structure proportions changed by less than 2 percent across the three time points within each sample, meaning the major conformational overhaul was essentially complete within the first hour of heating and then held steady. The loosened, open architecture of the high-dose phosphate samples aligns neatly with their higher free amino exposure and faster glycation kinetics.

To map exactly where glucose and its reactive descendants attached, the team turned to proteomics, digesting the glycated casein with trypsin and analyzing the resulting peptides by nano-scale liquid chromatography coupled to an Orbitrap mass spectrometer. They catalogued glycation sites, carboxymethylation, carboxyethylation, pyrraline, and the hydroimidazolone derivatives of glyoxal and methylglyoxal on lysine and arginine residues. After one hour, the MSP-treated samples showed markedly fewer detectable modification sites than the control, particularly for glucosylation and oxidation, while the DSP samples carried as many or more sites, especially for carboxymethylation and carboxyethylation. By four hours the picture shifted: MSP’s site counts rose steeply, suggesting its early inhibitory grip weakens with time, whereas some DSP markers declined, implying dynamic conversion or masking of modified residues. The authors caution that these counts reflect detectable site distribution rather than absolute glycation extent, since the analysis was not based on peptide-level quantification.

Perhaps the most commercially consequential findings concern flavor. An electronic nose fitted with ten sensors for compound classes ranging from aromatics to sulfides showed that after one hour, the control and high-dose DSP samples emitted the strongest volatile signals, while the MSP groups, especially the low-dose one, yielded the weakest. Gas chromatography-ion mobility spectrometry sharpened the picture: the untreated sample reached normalized intensity values of 100 for aldehydes and alcohols such as hexanal and 1-hexanol, evidence of amino acid degradation and Strecker degradation pathways, whereas high-dose MSP dramatically suppressed pentanal, 2-heptanone and ethyl propanoate. DSP-treated samples, meanwhile, maintained aldehyde, ketone and ester signals comparable to or above the control. Because the model system contained no lipids, these volatiles arose purely from protein-side chemistry, making the contrast especially clean.

Taken together, the study recasts phosphate salts as active, selective regulators of the Maillard reaction rather than inert processing aids. At high concentrations, both salts accelerate glycation by curbing aggregation, exposing amino groups and loosening protein structure, yet they diverge sharply downstream: MSP restrains early glycation and oxidative modification and suppresses aldehyde and ketone formation, supporting flavor stability, while DSP promotes browning and advanced glycation end product accumulation with little effect on the volatile profile. For manufacturers of processed cheese, milk powders and heat-treated dairy formulations, the message is that the choice between these two chemically similar salts is a choice between two different chemical futures for the product, one browner and more aromatically active, the other paler and more flavor-stable. The work offers a mechanistic roadmap for exploiting that divergence deliberately, turning a longstanding empirical art of dairy formulation into a matter of precise molecular control.

Subject of Research: Effects of sodium phosphate salts on the Maillard glycation of casein in dairy systems

Article Title: Effects of sodium dihydrogen phosphate (MSP) and disodium hydrogen phosphate (DSP) on the glycation of casein (CN)

Article References: Xu, S., Hu, X., Sun, Y., Huang, X., Zhou, Z., Han, L., Zhang, Y., Zhang, J., Gao, Y., Wang, L., & Qin, D. (2026). Effects of sodium dihydrogen phosphate (MSP) and disodium hydrogen phosphate (DSP) on the glycation of casein (CN). Food Chemistry: X, 39, Article 104621. https://doi.org/10.1016/j.fochx.2026.104621

Image Credits: AI Generated

DOI: 10.1016/j.fochx.2026.104621

Keywords: Maillard reaction, casein, glycation, phosphate salts, dairy processing, browning, food chemistry, protein structure, volatile compounds, flavor stability, processed cheese, mass spectrometry

The study, led by Xinyu Chen and Haoran Huang, focused on the later stages of hot-air drying, sampling marinated pork slices at 4, 4.5, 5, 5.5, and 6 hours, alongside an undried control. The researchers tracked moisture content, color, browning intensity, fluorescent Maillard reaction intermediates, and free amino acids across 180 sample units drawn from three biological replicates. Drying time exerted a highly significant effect on nearly every quality parameter measured, with statistical models confirming that the observed changes were driven by the drying process itself rather than by variation between procurement days or processing batches.

The chemical story that emerged is one of competing reactions unfolding at different speeds. As water evaporates, the food matrix contracts and flavor precursors become concentrated, accelerating lipid oxidation, Maillard reactions, and Strecker degradation. The team measured browning absorbance at 420 nanometers, which jumped from 16.82 in raw samples to 48.76 after just four hours and peaked at 67.80 at 5.5 hours. Fluorescence intensity, an early marker of Maillard progression, climbed steadily from roughly 6,000 to nearly 28,000 arbitrary units over the drying period, signaling the continuous accumulation of reactive intermediates even as the product approached its final low-moisture state.

Color told a parallel tale. Lightness values rose from 54.35 in raw meat to above 64 after five hours, a shift attributed to reduced surface moisture and thermal denaturation of myofibrillar proteins, both of which increase light scattering. Redness declined continuously as myoglobin degraded and oxidized, falling from 14.27 to 8.44 over six hours, while yellowness climbed from 21.18 to 27.71, reflecting the formation of Maillard intermediates and brown melanoidins. Together, these shifts describe jerky that becomes progressively lighter, less red, and more yellow-brown as drying proceeds, with the most dramatic changes occurring between four and five and a half hours.

Free amino acids, which contribute directly to taste and serve as precursors for aroma compounds, followed intricate trajectories. Glutamic acid, the dominant umami contributor, rose more than 50 percent to 1,286 milligrams per 100 grams of dry matter. Lysine surged more than twelvefold, while sweet-tasting alanine nearly doubled. In contrast, arginine collapsed from 443 to under 20 milligrams within the first four hours, and phenylalanine and tyrosine fell by roughly a third, likely consumed by Strecker degradation and related thermal reactions. The researchers note that the high initial arginine and aspartic acid levels partly reflect fish sauce in the marinade, a fermented seasoning naturally rich in free amino acids.

The volatile chemistry proved even more dynamic. Using gas chromatography-ion mobility spectrometry, or GC-IMS, the team identified 53 volatile organic compounds, including 19 aldehydes, 12 alcohols, 7 ketones, 8 heterocyclic compounds, 4 esters, 2 acids, and one miscellaneous compound. Crucially, individual compounds did not follow uniform trends. Pyrazines such as 2-ethyl-6-methylpyrazine spiked sharply after four hours and then declined, while furaneol dropped dramatically from its raw-sample level. Butanal dimers climbed steadily throughout, and furfuryl alcohol surged fifteenfold after drying began and then held relatively steady. Each measured signal represents a net balance between formation, volatilization, and thermal degradation, which is exactly why the researchers turned to machine learning to make sense of it.

The modeling strategy paired two fundamentally different algorithms. Support vector regression, which finds smooth nonlinear relationships in a kernel-defined feature space, and extra trees regression, which averages many randomized decision trees, were each trained to predict every physicochemical index directly from the 53 GC-IMS peak volumes. Both performed remarkably well, with test-set coefficients of determination between 0.989 and 0.998, but support vector regression was consistently more accurate, achieving a root mean square error of 0.602 for browning absorbance compared with 1.779 for extra trees. The similarity of results across training, validation, and test sets indicated no overfitting, demonstrating that the volatile fingerprints genuinely encode information about moisture, browning, and color.

High accuracy alone, however, says nothing about which compounds matter. To open the black box, the team applied SHAP, or SHapley Additive exPlanations, a framework that decomposes every prediction into contributions from individual variables. For moisture prediction, 1,2-dimethoxyethane and 1-pentanol dominated the support vector model, while heptanal and furfuryl alcohol led the extra trees version. Hexanal, a classic lipid oxidation product, ranked first for predicting both lightness and redness. The analysis also revealed that feature contributions shifted across drying stages, with many VOCs showing high importance in raw samples and different compounds taking over during the four-to-six-hour window, meaning no single global ranking fully captures the chemistry.

Importantly, several compounds appeared among the top predictors for both algorithms, including pentanal, furaneol, ethyl propanoate, 2-butanone, and 3-isobutyl-2-methoxypyrazine. Because their importance survived two very different modeling structures, the authors consider them comparatively robust candidate markers linking volatile profiles to quality attributes. The researchers are careful to emphasize that SHAP values explain model behavior and statistical associations within this dataset; they do not prove that these individual compounds directly participate in or cause the underlying reactions.

The sensory panel provided the human verdict, and it aligned neatly with the instrumental data. Scores for color, texture, and overall acceptability rose from four hours to a peak at five hours, where color reached 8.53, texture 8.63, and overall acceptability 8.43 on a nine-point scale, before declining at longer drying times. Aroma followed a similar pattern, though the difference between five and five and a half hours was not statistically significant. The convergence of sensory, physicochemical, and machine-learning evidence suggests a practical path forward: quality-control systems that read a jerky’s volatile fingerprint could verify drying endpoints without destructive testing. The authors caution that their models were built on a single raw-material source and drying range, so validation on independent production batches and earlier drying stages remains the next challenge before the approach reaches the factory floor.

Subject of Research: Effects of drying time on the physicochemical properties and volatile compounds of pork jerky analyzed by GC-IMS and interpretable machine learning

Article Title: Effects of drying time on physicochemical properties and volatile organic compounds of pork jerky revealed by GC-IMS and interpretable machine learning

Article References: Chen, X., Huang, H., Chen, H., Wang, S., Zang, Z., Huo, Y., Xiong, Z., Lu, D., & Li, Y. (2026). Effects of drying time on physicochemical properties and volatile organic compounds of pork jerky revealed by GC-IMS and interpretable machine learning. Food Chemistry: X, 39, Article 104562. https://doi.org/10.1016/j.fochx.2026.104562

Image Credits: AI Generated

DOI: 10.1016/j.fochx.2026.104562

Keywords: pork jerky, GC-IMS, machine learning, SHAP, Maillard reaction, drying time, volatile organic compounds, food chemistry, support vector regression, extra trees regression, free amino acids, sensory evaluation

The research team, led by Mei-Ling Li and Kuo-Chiang Hsu of China Medical University in Taiwan, chose pork for good reason: pork accounts for 32.8 percent of total meat intake in Taiwan, and charcoal-grilled pork is a staple of both household kitchens and street-food stalls. The investigators freeze-dried eight fresh fruits—guava, strawberry, lemon, kumquat, papaya, tankan orange, passion fruit, and banana—purchased from a local market in Taichung, milling the dried material into powders that could be standardized and dosed precisely. Four powders, guava, strawberry, lemon, and kumquat, were selected for meat experiments after screening showed they consistently outperformed the others in antioxidant capacity relative to their sugar content. The team then grilled pork loin and pork belly over a charcoal bed held at 500 ± 20 °C, monitored by infrared thermometry, until the geometric center of the meat reached 72 °C, the endpoint verified by thermocouple.

A key innovation of the study was treating the application method itself as a controllable processing variable rather than a fixed detail. The researchers compared three approaches: marinating, in which meat is immersed in a fruit powder solution for two hours at 4 °C; spraying, in which a calibrated atomizer delivers a mass-matched amount directly onto the surface; and mixing, in which ground patties are blended with an equivalent dry-matter loading of powder plus salt. Using a two-way analysis of variance to decompose the sources of variation in AGE inhibition, they found that in lean pork loin the application method accounted for a striking 62 percent of the total variance, compared with only 13 percent for fruit type and 17 percent for the interaction between the two. In other words, the way the protective formulation reached the meat mattered far more than the identity of the fruit itself.

The picture changed dramatically in the fattier pork belly, where the method contribution fell to 35 percent and the interaction term rose to 30 percent. Here the team uncovered a paradox that illustrates why sugar chemistry cannot be ignored: marinating pork belly with strawberry powder produced a negative inhibition rate of −142.5 percent, meaning AGE formation was dramatically promoted rather than suppressed. The likely culprit is the prolonged immersion, which allowed the strawberry powder’s high fructose and glucose loads—18.8 and 16.2 grams per 100 grams of dry powder, respectively—to penetrate the fat-rich intramuscular space and overwhelm the protective polyphenols delivered alongside them. Because fructose is eight to ten times more reactive than glucose in Maillard chemistry, sugar-rich fruit ingredients can actively backfire if they are given time and access to diffuse into the meat. Spraying, by contrast, kept inhibition above 60 percent for both kumquat (70.5 percent) and strawberry (60.9 percent) in belly, precisely because it confined the formulation to the surface without extended exposure.

These findings led the researchers to propose a new formulation-screening criterion: the antioxidant-to-reducing-sugar ratio, calculated as total phenolic or flavonoid content divided by the combined glucose and fructose burden. Guava, despite posting the highest ferric reducing antioxidant power (104.88 mg FeSO4 equivalents per gram) and the highest total phenolic content, carried the heaviest sugar load at 55.2 grams per 100 grams, yielding a phenolic-to-sugar ratio of just 0.35. Kumquat, with the highest flavonoid content (29.87 mg quercetin equivalents per gram) and comparatively low glucose, achieved the most favorable ratios. Correlation analysis reinforced the point: the phenolic-to-sugar ratios positively predicted AGE inhibition in belly-marinating conditions (r = 0.61 to 0.73), while FRAP alone actually showed a significant negative correlation with inhibition in loin-spraying conditions (r = −0.69). Conventional antioxidant metrics, the authors conclude, can be actively misleading when applied to fruit powder treatments in meat.

To translate the laboratory findings into something a home cook or food manufacturer could actually use, the team formulated two sprayable barbecue sauces: one containing soy sauce and one soy sauce-free, each fortified with the selected fruit powders, adjusted to pH 4.2–4.4, and hot-filled into 150-milliliter fine-mist spray bottles. The soy sauce choice proved consequential. The soy sauce-containing blank itself elevated total AGEs in pork belly by 58.2 percent over the untreated control, rising from 2.02 to 3.19 micrograms per gram, because the fermentation-derived free amino acids and reducing sugars in soy sauce serve as extra Maillard substrates during grilling. The soy sauce-free base, by contrast, actually reduced total AGEs by 27.4 percent in loin and 14.2 percent in belly on its own, providing a cleaner background against which the fruit polyphenols could work.

The final performance numbers were remarkable. In the soy sauce-free system, strawberry powder achieved the highest inhibition in pork loin, cutting total AGEs from 2.59 to 0.44 micrograms per gram—an 83.1 percent reduction versus the untreated control—with CML falling 98.2 percent, from 1.64 to 0.03 micrograms per gram. Kumquat powder delivered the best results in belly, reducing total AGEs from 2.02 to 0.28 micrograms per gram, an 86.2 percent reduction, with CML down 97.7 percent. The preferential suppression of CML over CEL suggests the sprays more efficiently blocked the oxidative glycation routes that produce CML than the methylglyoxal-associated pathways that feed CEL. Guava, despite its stellar antioxidant credentials, managed only 6.1 percent inhibition in the soy sauce-containing belly formulation, once again demonstrating that a high reducing-sugar load can negate antioxidant benefits even in a spray format.

Crucially, the optimized formulations passed the practical tests that determine whether a laboratory success can become a kitchen reality. Both strawberry and kumquat sprays significantly reduced lipid oxidation, cutting TBARS values by 53.0 and 32.3 percent respectively in loin, with all treated samples far below the 2.0 mg malondialdehyde per kilogram threshold associated with rancidity. A consumer panel of fifty tasters rated both sauces well above the neutral point on a nine-point hedonic scale, with strawberry earning an overall liking score of 6.33 and kumquat 6.14, and lemon was dropped from further development after panelists flagged pronounced bitterness from citrus limonoids. During accelerated storage at 45 °C for 90 days, both sauces retained more than 80 percent of their initial flavonoid content and kept AGE inhibition rates between roughly 51 and 84 percent depending on matrix and powder, while triangle tests found the flavor indistinguishable from the day-zero reference throughout the entire storage period.

To explain why the sprays work, the researchers combined untargeted metabolomics with computational and biochemical probes. Twenty-eight polyphenols identified in the kumquat and strawberry powders were docked against four glycation-susceptible hotspots on porcine myosin, using an AlphaFold-predicted protein structure; 27 of the 28 bound more strongly than glucose, and all 28 outcompeted the reactive carbonyl intermediates methylglyoxal and glyoxal, with top performers including hesperidin, neohesperidin, and procyanidin B2. Methylglyoxal-trapping assays then revealed a clear structure–activity relationship: dihydromyricetin, with three free hydroxyl groups on its B-ring, trapped 85.2 percent of methylglyoxal, while fully methoxylated nobiletin trapped a negligible 2.1 percent despite moderate docking affinity. The authors propose a three-pathway model in which polyphenols competitively occupy glycation sites on myosin, directly trap reactive dicarbonyl intermediates, and suppress lipid oxidation that would otherwise generate secondary reactive carbonyls. While the docking and trapping data do not constitute direct proof inside the grilled-meat matrix, they offer a coherent molecular framework for a deceptively simple message: when it comes to healthier grilling, a fine mist of the right fruit powder, applied at the right moment, may matter as much as what is in the bottle.

Subject of Research: Reducing advanced glycation end-product formation in charcoal-grilled pork using spray-applied fruit powder formulations

Article Title: Fruit powder spray formulations reduce advanced glycation end-product formation in charcoal-grilled pork: effects of application method and polyphenol–sugar balance

Article References: Li, M.-L., Huang, S.-M., Wu, P.-Y., Huang, C.-W., Hung, W.-L., & Hsu, K.-C. (2026). Fruit powder spray formulations reduce advanced glycation end-product formation in charcoal-grilled pork: effects of application method and polyphenol–sugar balance. Current Research in Food Science, 13, Article 101592. https://doi.org/10.1016/j.crfs.2026.101592

Image Credits: AI Generated

DOI: 10.1016/j.crfs.2026.101592

Keywords: advanced glycation end-products, charcoal grilling, fruit powder, polyphenols, Maillard reaction, pork, food chemistry, spray application, CML, CEL, methylglyoxal, barbecue sauce

Peanuts, Arachis hypogaea, are one of the world’s three major oilseed crops, and their roasted flavor carries enormous commercial weight. Yet despite their importance, the molecular pathways and interactions that generate flavor during roasting have never been systematically established. The problem is one of scale and simultaneity. Roasting triggers Maillard reactions between amino acids and sugars, lipid oxidation and degradation, caramelization, and a cascade of secondary reactions, all unfolding in parallel inside a heterogeneous matrix of cells, lipid droplets, and protein bodies. Traditional flavor chemistry has typically tracked a handful of volatile compounds over time, but it has rarely connected those measurements back to the specific precursor molecules and the physical transformations of the tissue that release them.

The new study tackles that gap by integrating what the authors describe as food multi-scale structural theory with food multi-omics technology. In practical terms, the team monitored how the visible appearance and the cellular microstructure of peanuts changed as roasting progressed, while simultaneously profiling the lipids, amino acids, and oligosaccharides inside the kernels and the volatile flavor compounds that emerged. By layering these datasets on top of one another, they could correlate specific precursor molecules with specific flavor products, and anchor both to defined stages of the roasting process. The result is a proposed model of the molecular mechanisms of peanut flavor formation, a framework the authors say can serve as a foundation for designing oilseed products with enhanced flavor profiles.

One of the study’s most striking contributions is quantitative. The researchers found that changes in the color spectrum of the roasting peanuts could be described by twelve quadratic polynomial fitting models, all with coefficients of determination greater than 0.88 and the highest reaching 0.98. In other words, the visible browning of a peanut, long used informally by roasters as a proxy for doneness, follows a mathematically predictable trajectory that tracks the underlying chemistry. Because color development during roasting is largely a readout of Maillard reaction progress, these models offer a non-destructive way to estimate how far the flavor-generating reactions have advanced simply by measuring the surface color of the kernels.

The microstructural side of the analysis proved equally revealing. Using observations of cellular architecture across the roasting timeline, the team constructed a spatial model of microstructural change that they characterize as a lipid droplet collision-outward diffusion process. As heat penetrates the kernel, lipid droplets within the cells move, collide, and their contents diffuse outward through the tissue. This physical mobilization matters because the lipid fraction of the peanut is not merely a passive reservoir of fatty acids; lipid degradation products contribute directly to roasted aroma, and the redistribution of lipids can alter how precursor molecules encounter one another. The spatial model thus links the microscopic rearrangement of the kernel’s interior to the chemical availability of flavor precursors.

Bringing the chemical and structural data together, the researchers identified three distinct phases of the roasting process at approximately 160 degrees Celsius. The first, spanning roughly the first five minutes, is a heat-induced mass-transfer phase, during which water migrates and molecules begin to move within the tissue. The second, from five to ten minutes, is a heat-induced activation phase, in which the reactants become chemically mobilized and the conditions for flavor-generating reactions are established. The third, at around fifteen minutes, is a heat-responsive reaction phase, when the characteristic flavor compounds, particularly the pyrazines that give roasted peanuts their signature nutty note, are actively produced. This phase framework transforms roasting from an opaque artisanal process into a staged, mechanistically defined sequence.

The multi-omics analysis itself pinpointed the key molecular players with unusual precision. The team identified the seven most prominent lipid molecules, six amino acid molecules, and seven oligosaccharide molecules involved in peanut flavor formation, along with the key flavor molecules to which they give rise. Pyrazines, heterocyclic nitrogen-containing compounds formed primarily through Maillard chemistry, emerged as the central targets of the analysis, consistent with their status as the dominant contributors to roasted peanut aroma. By establishing statistical correlations between the identified precursor substances and the characteristic flavor compounds, the researchers could propose which amino acids and sugars feed into which aroma products, and at which stage of the roasting process those conversions become significant.

What makes this study notable beyond the peanut itself is its methodological template. The combination of multi-scale structural observation with multi-omics profiling, wrapped in quantitative models, offers a blueprint that could be applied to other oilseeds and roasted foods, from sesame and sunflower seeds to coffee and cocoa. Flavor formation in all of these systems involves the same broad families of reactions, yet each food matrix imposes its own physical constraints on how precursors are released and mixed. A framework that explicitly connects microstructure, precursor chemistry, and flavor output addresses a long-standing weakness in food science, where volatile profiles are often catalogued in detail but rarely explained mechanistically from the ground up.

For the food industry, the practical implications are immediate. Roasting schedules for peanuts and peanut products are typically optimized empirically, with manufacturers tuning time and temperature by trial and error to hit desired sensory targets. A validated mechanistic model changes that calculus. If the three-phase structure of roasting is generalizable, processors could design heating profiles that deliberately manage the mass-transfer, activation, and reaction phases separately, for example by controlling moisture migration early and modulating temperature precisely when pyrazine formation peaks. The color-based polynomial models could support real-time process monitoring, allowing production lines to track flavor development non-destructively through optical measurements rather than relying solely on timed endpoints or sensory panels.

There are also implications for breeding and raw material selection. Knowing which specific lipid, amino acid, and oligosaccharide molecules drive flavor formation means that the flavor potential of a peanut variety could in principle be assessed, or even selected for, at the molecular level before a single kernel is roasted. That opens a path toward crop improvement programs aimed at flavor, a trait that has historically been difficult to quantify and even harder to predict. The authors frame their proposed molecular mechanism model as a foundation for developing oilseed products with enhanced flavor profiles, and the identification of a defined set of precursor molecules gives breeders and food technologists a concrete molecular checklist to work from. As the global demand for plant-based proteins and flavorful oilseed products continues to grow, understanding exactly how heat transforms a raw peanut into one of the world’s most comforting aromas is no longer just a matter of curiosity. It is becoming an engineering problem, and this study provides some of the first rigorous equations for solving it.

Subject of Research: Molecular mechanisms of flavor compound formation during peanut roasting

Article Title: Elucidating flavor formation molecular mechanisms in peanut roasting through multi-omics analysis

Article References: Xiang, F., Hu, H., Ding, C., Ma, X., Xu, H., Jiang, Y., Shi, H., Adhikari, B., Xu, X., Shi, A., & Wang, Q. (2026). Elucidating flavor formation molecular mechanisms in peanut roasting through multi-omics analysis. npj Science of Food. https://doi.org/10.1038/s41538-026-01127-5

Image Credits: AI Generated

DOI: 10.1038/s41538-026-01127-5

Keywords: peanut roasting, flavor chemistry, pyrazines, Maillard reaction, multi-omics, lipidomics, amino acids, oligosaccharides, food microstructure, oilseeds, roasting phases, food science

The study, published in Current Research in Food Science, comes at a time when the World Health Organization continues to press for population-wide sodium reduction, and bread remains one of the largest dietary contributors simply because people eat so much of it. Previous attempts to swap in potassium chloride often failed on flavor, leaving a bitter or metallic aftertaste that consumers reject. Yeast extract, a natural ingredient rich in amino acids, peptides, and nucleotides, has long been suspected of a cleverer role: rather than mimicking salt’s chemistry, it could compensate for the lost saltiness through umami, the savory fifth taste.

What makes the new work unusual is its completeness. Earlier studies had tested yeast extract in crackers or spelt bread, or combined it with potassium chloride, but none had replaced sodium chloride across a full gradient in a single product while tracking dough rheology, molecular structure, texture, color, sensory scores, and microbial counts. The researchers baked baguettes in which 0, 25, 50, 75, or 100 percent of the salt was replaced by commercial yeast extract, then froze the loaves and followed them for 90 days.

The rheological measurements told a story of progressive weakening. In farinograph tests, dough stability time fell from just over seven minutes in the fully salted control to barely three minutes when all the salt was replaced, while the degree of softening more than doubled. Extensograph analysis showed the dough becoming dramatically more stretchable and less resistant to deformation, with the resistance-to-extensibility ratio dropping from 4.22 to 1.87. The explanation is electrostatic: salt ions shield charges on gluten proteins, allowing them to pack into a strong, cohesive network. Remove the ions, and the network loosens. Yeast extract, being organic rather than ionic, simply cannot fill that structural role.

Yet the chemistry of the bread itself changed in intriguing ways. Protein content climbed steadily with yeast extract, reaching nearly 13 percent at full replacement, a direct nutritional bonus from the extract’s nitrogen-rich compounds. Sodium content, meanwhile, plummeted from 0.42 percent in the control to just 0.07 percent in the fully substituted loaf, a reduction of more than 80 percent. Infrared spectroscopy revealed intensified amide bands confirming the added protein, and showed that yeast extract strengthened the short-range ordered structure of starch, a molecular feature associated with slower digestion by enzymes. X-ray diffraction detected a new crystalline peak in the high-replacement breads, hinting at altered starch organization during fermentation.

Color analysis brought an unexpected consumer-friendly finding. While crumb color stayed essentially unchanged, the crust became significantly redder and more yellow as yeast extract levels rose. The reason is the Maillard reaction: yeast extract is loaded with free amino acids that react vigorously with reducing sugars under baking heat, producing the red-brown melanoidin pigments that bakers and shoppers alike associate with a well-baked, flavorful loaf. In other words, the salt substitute made the bread look more appetizing, not less.

Sensory results were equally encouraging. A trained panel of thirteen assessors, blinded to formulation, found no statistically significant differences among the five breads in color, odor, texture, or overall acceptability. Only flavor showed a measurable decline, and even then only the fully salt-free bread scored clearly below the control, dropping from 4.17 to 3.33 on a five-point scale. The glutamates and nucleotides in yeast extract appear to have done exactly what the umami hypothesis predicted, masking much of the sensory deficit left by the missing salt across a wide range of replacement levels.

Texture, however, exposed the trade-off. Crumb hardness rose consistently with replacement level, from 46 grams in the control to nearly 82 grams at full substitution on baking day, and the gap widened after 90 days of frozen storage, when the fully replaced bread reached 128 grams. Chewiness followed the same pattern. The researchers attribute this to altered water distribution and protein-starch interactions that accelerate the recrystallization of amylopectin, the molecular engine of staling. Frozen storage itself also damaged the crumb, as growing ice crystals pressed against the structure and reduced cohesiveness and springiness in every loaf regardless of formulation.

Microbiology delivered the study’s most sobering lesson. Total bacterial counts on production day climbed in lockstep with yeast extract levels, from about 117 colony-forming units per gram in the control to 760 in the fully replaced bread. Commercial yeast extract is not sterile, and heat-resistant spores evidently survived baking. No yeasts or molds were detected in any sample, and freezing actually reduced bacterial counts across the board over 90 days, but the finding underscores that a biologically derived salt replacer carries a microbial load that pure sodium chloride never does. The authors suggest pre-treatments such as thermal processing or irradiation of the extract before use.

Balancing all of the evidence, the researchers conclude that replacing 25 to 50 percent of salt with yeast extract offers the best compromise: enough sodium chloride remains to support gluten strength and microbial control, while the extract contributes umami flavor, extra protein, and an appealingly browned crust with minimal microbial burden. Higher replacement levels, though sensorially tolerable in a trained panel, weakened the dough, firmed the crumb, and raised the initial bacterial load. The team cautions that larger consumer panels, industrial-scale trials, and longer storage studies are still needed before salt-reduced baguettes built on yeast extract reach the bakery shelf, but the blueprint for a tastier, healthier loaf is now on the table.

Subject of Research: Yeast extract as a sodium chloride replacer in baguette bread and its effects on dough rheology, bread quality, and frozen storage stability

Article Title: Yeast Extract for Salt Reduction in Baguette Bread: A Comprehensive Evaluation of Dough Properties, Bread Quality, and Storage Stability

Article References: Hedayati, S., Chaychi, F., Babaali, E., & Kazemi, A. (2026). Yeast extract for salt reduction in baguette bread: A comprehensive evaluation of dough properties, bread quality, and storage stability. Current Research in Food Science, 13, Article 101590. https://doi.org/10.1016/j.crfs.2026.101590

Image Credits: AI Generated

DOI: 10.1016/j.crfs.2026.101590

Keywords: yeast extract, salt reduction, sodium, baguette, bread quality, dough rheology, gluten, umami, frozen storage, food science, Maillard reaction, food microbiology

Keemun black tea, produced from the Camellia sinensis cultivar Qimen in Anhui province, is prized for its pronounced floral, fruity and sweet notes, a profile that has earned it a place among the most famous black teas on Earth. Its manufacture follows a strict sequence of withering, rolling, fermentation and drying. While withering and fermentation have attracted considerable research attention, the drying step, despite being the final thermal process that locks in color, taste and aroma, has remained surprisingly underexplored. Previous studies largely focused on drying temperature or variable temperature profiles, leaving the role of drying time itself largely unquantified.

To close that gap, the research team, led by Erdenechimeg Otgonbat and colleagues, sampled tea directly from a commercial-scale hot-air dryer operating at 110 degrees Celsius. Subsamples of roughly 100 grams were pulled from three separate grids of the drying bed before drying and after 10, 15, 20, 25 and 30 minutes, providing biological replicates that eliminated micro-environmental gradients inside the machine. Crucially, every sample was immediately plunged into liquid nitrogen and freeze-dried. This lyophilization step served two purposes: it standardized all samples to a uniform dry weight, because headspace partitioning of volatiles is exquisitely sensitive to moisture, and it halted residual enzymatic activity and thermal reactions, preserving a genuine chemical snapshot of each moment.

The volatiles were then captured using headspace solid-phase microextraction, a solvent-free technique, followed by gas chromatography-mass spectrometry. In total, the team identified 104 volatile compounds spanning alcohols, aldehydes, esters, alkenes, ketones, alkanes and heterocyclics. Alcohols dominated the profile, accounting for between 66 and 71 percent of total volatiles, with aldehydes close behind. The total volatile content followed a strikingly non-linear trajectory: it fell from 18,148 micrograms per kilogram before drying to a low of 14,623 at 15 minutes, then surged to 16,856 at 20 minutes before settling near 15,679 at the end of the process.

That transient spike at 20 minutes is one of the study’s most intriguing observations. The researchers hypothesize that it reflects a dynamic equilibrium between physical evaporation, which strips volatiles away in the hot-air stream, and thermal chemical generation, which creates new aroma molecules. As heat accumulates, non-enzymatic cascades such as the Maillard reaction and Strecker degradation of amino acids churn out fresh aldehydes and heterocycles, while thermal hydrolysis cleaves non-volatile glycosidically bound precursors, liberating a wave of free linalool and geraniol derivatives. For a brief window, the rate of generation outpaces the rate of loss. Beyond 20 minutes, precursors become depleted and the balance tips back toward evaporation. The authors caution that this mechanism remains a working hypothesis requiring isotope-labeling studies to confirm.

Sensory evaluation, conducted under China’s national standard for tea assessment with a panel of seven trained assessors, corroborated the chemistry. As drying progressed, the infusion color deepened from bright orange-yellow to darker orange-red, while green and astringent notes faded completely between 20 and 30 minutes. By the end of drying, the tea reached its peak sensory quality, displaying the classic high sweet and floral Keemun aroma alongside a bright red liquor. The phenotypic trajectory of the panel aligned neatly with the molecular data, giving the chemical findings a human anchor.

To identify which compounds actually mattered, the team deployed partial least squares discriminant analysis, validated with a 999-permutation test to rule out overfitting. This chemometric screen flagged 20 differential volatile markers that reliably distinguished the drying intervals. A subsequent odor activity value analysis, which compares each compound’s concentration against its odor threshold in water, narrowed the field to 39 aroma-active compounds, of which six emerged as the primary drivers of Keemun’s signature scent: trans-beta-ionone, linalool, geraniol, benzeneacetaldehyde, nonanal and theaspirane.

The dynamics of these key odorants reveal two competing chemical stories. Trans-beta-ionone and related C13-norisoprenoids surged during the 20-to-30-minute window, direct evidence of thermal oxidative cleavage of carotenoid pigments such as beta-carotene and neoxanthin, which fragment into sweet-floral and sweet-woody molecules once thermal energy overcomes the activation barrier under low-moisture conditions. Meanwhile, the monoterpene alcohols linalool and geraniol fluctuated as residual glycoside-hydrolyzing enzymes released them early in drying, only for desiccation to inactivate those enzymes past 20 minutes, letting volatilization win. Most tellingly, benzeneacetaldehyde, the honey-like note, accumulated late in drying through Strecker degradation of the amino acid phenylalanine within the Maillard reaction cascade, confirming that drying acts not merely as dehydration but as a thermal-chemical reactor.

The practical implications extend beyond academic curiosity. Because the most critical accumulation of Keemun’s key odorants occurs in the middle-to-late drying stages, manufacturers now have a quantitative basis for tuning drying time as a precision parameter alongside temperature, potentially enabling smarter, aroma-targeted control in tea production. The authors acknowledge limitations, noting that absolute quantification could be cross-validated with solvent-assisted flavor evaporation and that future multi-omics work should trace precursor degradation kinetics. Still, the study delivers something rare in food science: a minute-by-minute molecular biography of a flavor in the making, showing exactly when a leaf stops smelling green and starts smelling like Keemun.

Subject of Research: Time-dependent changes in volatile and aroma compounds during the hot-air drying of Keemun black tea

Article Title: Time-dependent changes of aroma and volatile compounds during drying of Keemun black tea

Article References: Otgonbat, E., Cui, J., Luo, Z., Song, Y., Gao, T., & Song, C. (2026). Time-dependent changes of aroma and volatile compounds during drying of Keemun black tea. Food Chemistry: X, Article 104531. https://doi.org/10.1016/j.fochx.2026.104531

Image Credits: AI Generated

DOI: 10.1016/j.fochx.2026.104531

Keywords: Keemun black tea, tea drying, aroma chemistry, volatile compounds, Maillard reaction, Strecker degradation, HS-SPME-GC/MS, linalool, trans-beta-ionone, odor activity value, food chemistry, sensory evaluation